> I think most of these are because in the past (yes I have fixed that!!) > all 'sizetype' constants were sign-extended (and the signedness, > that is, TYPE_UNSIGNED (sizetype), was frontend dependend (ugh) and > then later true, thus unsigned). > > So I think all _SIZE stuff should check fits_uhwi_p and be used as > uhwi. But that may have ripple-down effects, so consistently > using fits_shwi_p and using as shwi is also fine (it just restricts > the maximum values we accept(?)).
And please make sure to test Ada if you're tweaking this, it's fragile stuff. -- Eric Botcazou
